We present SOPHY, a generative model for 3D physics-aware shape synthesis. Unlike existing 3D generative models that focus solely on static geometry or 4D models that produce physics-agnostic animations, our method jointly synthesizes shape, texture, and material properties related to physics-grounded dynamics, making the generated objects ready for simulations and interactive, dynamic environments. To train our model, we introduce a dataset of 3D objects annotated with detailed physical material attributes, along with an efficient pipeline for material annotation. Our method enables applications such as text-driven generation of interactive, physics-aware 3D objects and single-image reconstruction of physically plausible shapes. Furthermore, our experiments show that jointly modeling shape and material properties enhances the realism and fidelity of the generated shapes, improving performance on both generative geometry and physical plausibility. Project page: https://xjay18.github.io/SOPHY.

We present mean-shift distillation, a novel diffusion distillation technique that provides a provably good proxy for the gradient of the diffusion output distribution. This is derived directly from mean-shift mode seeking on the distribution, and we show that its extrema are aligned with the modes. We further derive an efficient product distribution sampling procedure to evaluate the gradient. Our method is formulated as a drop-in replacement for score distillation sampling (SDS), requiring neither model retraining nor extensive modification of the sampling procedure. We show that it exhibits superior mode alignment as well as improved convergence in both synthetic and practical setups, yielding higher-fidelity results when applied to both text-to-image and text-to-3D applications with Stable Diffusion.

We introduce ShapeWords1, an approach for synthesizing images based on 3D shape guidance and text prompts. ShapeWords incorporates target 3D shape information within specialized tokens embedded together with the input text, effectively blending 3D shape awareness with textual context to guide the image synthesis process. Unlike conventional shape guidance methods that rely on depth maps restricted to fixed viewpoints and often overlook full 3D structure or textual context, ShapeWords generates diverse yet consistent images that reflect both the target shape’s geometry and the textual description. Experimental results show that ShapeWords produces images that are more text-compliant, aesthetically plausible, while also maintaining 3D shape awareness.

We present GEOPARD, a transformer-based architecture for predicting articulation from a single static snapshot of a 3D shape. The key idea of our method is a pretraining strategy that allows our transformer to learn plausible candidate articulations for 3D shapes based on a geometric-driven search without manual articulation annotation. The search automatically discovers physically valid part motions that do not cause detachments or collisions with other shape parts. Our experiments indicate that this geometric pretraining strategy, along with carefully designed choices in our transformer architecture, yields state-of-the-art results in articulation inference in the PartNet-Mobility dataset.

Tracking dense 3D motion from monocular videos remains challenging, particularly when aiming for pixel-level precision over long sequences. We introduce DELTA, a novel method that efficiently tracks every pixel in 3D space, enabling accurate motion estimation across entire videos. Our approach leverages a joint global-local attention mechanism for reduced-resolution tracking, followed by a transformer-based upsampler to achieve high-resolution predictions. Unlike existing methods, which are limited by computational inefficiency or sparse tracking, DELTA delivers dense 3D tracking at scale, running over 8x faster than previous methods while achieving state-of-the-art accuracy. Furthermore, we explore the impact of depth representation on tracking performance and identify log-depth as the optimal choice. Extensive experiments demonstrate the superiority of DELTA on multiple benchmarks, achieving new state-of-the-art results in both 2D and 3D dense tracking tasks. Our method provides a robust solution for applications requiring fine-grained, long-term motion tracking in 3D space.

Abstract We present Im2SurfTex, a method that generates textures for input 3D shapes by learning to aggregate multi‐view image outputs produced by 2D image diffusion models onto the shapes' texture space. Unlike existing texture generation techniques that use ad hoc backprojection and averaging schemes to blend multiview images into textures, often resulting in texture seams and artifacts, our approach employs a trained neural module to boost texture coherency. The key ingredient of our module is to leverage neural attention and appropriate positional encodings of image pixels based on their corresponding 3D point positions, normals, and surface‐aware coordinates as encoded in geodesic distances within surface patches. These encodings capture texture correlations between neighboring surface points, ensuring better texture continuity. Experimental results show that our module improves texture quality, achieving superior performance in high‐resolution texture generation.

We introduce Open3DIS, a novel solution designed to tackle the problem of Open-Vocabulary Instance Segmentation within 3D scenes. Objects within 3D environments exhibit diverse shapes, scales, and colors, making precise instance-level identification a challenging task. Recent advancements in Open-Vocabulary scene understanding have made significant strides in this area by employing class-agnostic 3D instance proposal networks for object localization and learning queryable features for each 3D mask. While these methods produce high-quality instance proposals, they struggle with identifying small-scale and geometrically ambiguous objects. The key idea of our method is a new module that aggregates 2D instance masks across frames and maps them to geometrically coherent point cloud regions as high-quality object proposals addressing the above limitations. These are then combined with 3D class-agnostic instance proposals to include a wide range of objects in the real world. To validate our approach, we conducted experiments on three prominent datasets, including ScanNet200, S3DIS, and Replica, demonstrating significant performance gains in segmenting objects with diverse categories over the state-of-the-art approaches.

We present VecFusion, a new neural architecture that can generate vector fonts with varying topological structures and precise control point positions. Our approach is a cascaded diffusion model which consists of a raster diffusion model followed by a vector diffusion model. The raster model generates low-resolution, rasterized fonts with auxiliary control point information, capturing the global style and shape of the font, while the vector model synthesizes vector fonts conditioned on the low-resolution raster fonts from the first stage. To synthesize long and complex curves, our vector diffusion model uses a transformer architecture and a novel vector representation that enables the modeling of diverse vector geometry and the precise prediction of control points. Our experiments show that, in contrast to previous generative models for vector graphics, our new cascaded vector diffusion model generates higher quality vector fonts, with complex structures and diverse styles.

The success of denoising diffusion models in representing rich data distributions over 2D raster images has prompted research on extending them to other data representations, such as vector graphics. Unfortunately due to their variable structure and scarcity of vector training data, directly applying diffusion models on this domain remains a challenging problem. Using workarounds like optimization via Score Distillation Sampling (SDS) is also fraught with difficulty, as vector representations are non-trivial to directly optimize and tend to result in implausible geometries such as redundant or self-intersecting shapes. NIVeL addresses these challenges by reinterpreting the problem on an alternative, intermediate domain which preserves the desirable properties of vector graphicsmainly sparsity of representation and resolution-independence. This alternative domain is based on neural implicit fields expressed in a set of decomposable, editable layers. Based on our experiments, NIVeL produces text-to- vector graphics results of significantly better quality than the state-of-the-art.

We present ANISE, a method that reconstructs a 3D shape from partial observations (images or sparse point clouds) using a part-aware neural implicit shape representation. The shape is formulated as an assembly of neural implicit functions, each representing a different part instance. In contrast to previous approaches, the prediction of this representation proceeds in a coarse-to-fine manner. Our model first reconstructs a structural arrangement of the shape in the form of geometric transformations of its part instances. Conditioned on them, the model predicts part latent codes encoding their surface geometry. Reconstructions can be obtained in two ways: (i) by directly decoding the part latent codes to part implicit functions, then combining them into the final shape; or (ii) by using part latents to retrieve similar part instances in a part database and assembling them in a single shape. We demonstrate that, when performing reconstruction by decoding part representations into implicit functions, our method achieves state-of-the-art part-aware reconstruction results from both images and sparse point clouds. When reconstructing shapes by assembling parts retrieved from a dataset, our approach significantly outperforms traditional shape retrieval methods even when significantly restricting the database size. We present our results in well-known sparse point cloud reconstruction and single-view reconstruction benchmarks.

Abstract We present a deep learning method that propagates point‐wise feature representations across shapes within a collection for the purpose of 3D shape segmentation. We propose a cross‐shape attention mechanism to enable interactions between a shape's point‐wise features and those of other shapes. The mechanism assesses both the degree of interaction between points and also mediates feature propagation across shapes, improving the accuracy and consistency of the resulting point‐wise feature representations for shape segmentation. Our method also proposes a shape retrieval measure to select suitable shapes for cross‐shape attention operations for each test shape. Our experiments demonstrate that our approach yields state‐of‐the‐art results in the popular PartNet dataset.

Rigged puppets are one of the most prevalent representations to create 2D character animations. Creating these puppets requires partitioning characters into independently moving parts. In this work, we present a method to automatically identify such articulated parts from a small set of character poses shown in a sprite sheet, which is an illustration of the character that artists often draw before puppet creation. Our method is trained to infer articulated parts, e.g. head, torso and limbs, that can be reassembled to best reconstruct the given poses. Our results demonstrate significantly better performance than alternatives qualitatively and quantitatively. Our project page https://zhan-xu.github.io/parts/ includes our code and data.

Human speech is often accompanied by body gestures including arm and hand gestures. We present a method that reenacts a high-quality video with gestures matching a target speech audio. The key idea of our method is to split and re-assemble clips from a reference video through a novel video motion graph encoding valid transitions between clips. To seamlessly connect different clips in the reenactment, we propose a pose-aware video blending network which synthesizes video frames around the stitched frames between two clips. Moreover, we developed an audio-based gesture searching algorithm to find the optimal order of the reenacted frames. Our system generates reen-actments that are consistent with both the audio rhythms and the speech content. We evaluate our synthesized video quality quantitatively, qualitatively, and with user studies, demonstrating that our method produces videos of much higher quality and consistency with the target audio compared to previous work and baselines. Our project page https://github.com/yzhou359/vid-reenact includes code and data.

We present MoRig, a method that automatically rigs character meshes driven by single-view point cloud streams capturing the motion of performing characters. Our method is also able to animate the 3D meshes according to the captured point cloud motion. MoRig’s neural network encodes motion cues from the point clouds into features that are informative about the articulated parts of the performing character. These motion-aware features guide the inference of an appropriate skeletal rig for the input mesh, which is then animated based on the point cloud motion. Our method can rig and animate diverse characters, including humanoids, quadrupeds, and toys with varying articulation. It accounts for occluded regions in the point clouds and mismatches in the part proportions between the input mesh and captured character. Compared to other rigging approaches that ignore motion cues, MoRig produces more accurate rigs, well-suited for re-targeting motion from captured characters.

Abstract We present PriFit, a semi‐supervised approach for label‐efficient learning of 3D point cloud segmentation networks. PriFit combines geometric primitive fitting with point‐based representation learning. Its key idea is to learn point representations whose clustering reveals shape regions that can be approximated well by basic geometric primitives, such as cuboids and ellipsoids. The learned point representations can then be re‐used in existing network architectures for 3D point cloud segmentation, and improves their performance in the few‐shot setting. According to our experiments on the widely used ShapeNet and PartNet benchmarks, PriFit outperforms several state‐of‐the‐art methods in this setting, suggesting that decomposability into primitives is a useful prior for learning representations predictive of semantic parts. We present a number of ablative experiments varying the choice of geometric primitives and downstream tasks to demonstrate the effectiveness of the method.

Constructive solid geometry (CSG) is a geometric modeling technique that defines complex shapes by recursively applying boolean operations on primitives such as spheres and cylinders. We present CSGNet, a deep network architecture that takes as input a 2D or 3D shape and outputs a CSG program that models it. Parsing shapes into CSG programs is desirable as it yields a compact and interpretable generative model. However, the task is challenging since the space of primitives and their combinations can be prohibitively large. CSGNet uses a convolutional encoder and recurrent decoder based on deep networks to map shapes to modeling instructions in a feed-forward manner and is significantly faster than bottom-up approaches. We investigate two architectures for this task-a vanilla encoder (CNN) - decoder (RNN) and another architecture that augments the encoder with an explicit memory module based on the program execution stack. The stack augmentation improves the reconstruction quality of the generated shape and learning efficiency. Our approach is also more effective as a shape primitive detector compared to a state-of-the-art object detector. Finally, we demonstrate CSGNet can be trained on novel datasets without program annotations through policy gradient techniques.

We present ASSET, a neural architecture for automatically modifying an input high-resolution image according to a user's edits on its semantic segmentation map. Our architecture is based on a transformer with a novel attention mechanism. Our key idea is to sparsify the transformer's attention matrix at high resolutions, guided by dense attention extracted at lower image resolutions. While previous attention mechanisms are computationally too expensive for handling high-resolution images or are overly constrained within specific image regions hampering long-range interactions, our novel attention mechanism is both computationally efficient and effective. Our sparsified attention mechanism is able to capture long-range interactions and context, leading to synthesizing interesting phenomena in scenes, such as reflections of landscapes onto water or fora consistent with the rest of the landscape, that were not possible to generate reliably with previous convnets and transformer approaches. We present qualitative and quantitative results, along with user studies, demonstrating the effectiveness of our method. Our code and dataset are available at our project page: https://github.com/DifanLiu/ASSET

This paper proposes a method for automatic generation of textures for 3D city meshes in immersive urban environments. Many recent pipelines capture or synthesize large quantities of city geometry using scanners or procedural modeling pipelines. Such geometry is intricate and realistic, however the generation of photo-realistic textures for such large scenes remains a problem. We propose to generate textures for input target 3D meshes driven by the textural style present in readily available datasets of panoramic photos capturing urban environments. Re-targeting such 2D datasets to 3D geometry is challenging because the underlying shape, size, and layout of the urban structures in the photos do not correspond to the ones in the target meshes. Photos also often have objects (e.g., trees, vehicles) that may not even be present in the target geometry. To address these issues we present a method, called Projective Urban Texturing (PUT), which re-targets textural style from real-world panoramic images to unseen urban meshes. PUT relies on contrastive and adversarial training of a neural architecture designed for unpaired image-to-texture translation. The generated textures are stored in a texture atlas applied to the target 3D mesh geometry. To promote texture consistency, PUT employs an iterative procedure in which texture synthesis is conditioned on previously generated, adjacent textures. We demonstrate both quantitative and qualitative evaluation of the generated textures.

This paper introduces a model for producing stylized line drawings from 3D shapes. The model takes a 3D shape and a viewpoint as input, and outputs a drawing with textured strokes, with variations in stroke thickness, deformation, and color learned from an artist’s style. The model is fully differentiable. We train its parameters from a single training drawing of another 3D shape. We show that, in contrast to previous image-based methods, the use of a geometric representation of 3D shape and 2D strokes allows the model to transfer important aspects of shape and texture style while preserving contours. Our method outputs the resulting drawing in a vector representation, enabling richer downstream analysis or editing in interactive applications. Our code and dataset are available at our project page: www.github.com/DifanLiu/NeuralStrokes

We introduce BuildingNet: (a) a large-scale dataset of 3D building models whose exteriors are consistently labeled, and (b) a graph neural network that labels building meshes by analyzing spatial and structural relations of their geometric primitives. To create our dataset, we used crowdsourcing combined with expert guidance, resulting in 513K annotated mesh primitives, grouped into 292K semantic part components across 2K building models. The dataset covers several building categories, such as houses, churches, skyscrapers, town halls, libraries, and castles. We include a benchmark for evaluating mesh and point cloud labeling. Buildings have more challenging structural complexity compared to objects in existing benchmarks (e.g., ShapeNet, PartNet), thus, we hope that our dataset can nurture the development of algorithms that are able to cope with such large-scale geometric data for both vision and graphics tasks e.g., 3D semantic segmentation, part-based generative models, correspondences, texturing, and analysis of point cloud data acquired from real-world buildings. Finally, we show that our mesh-based graph neural network significantly improves performance over several baselines for labeling 3D meshes. Our project page www.buildingnet.org includes our dataset and code.

This paper introduces a method for learning to generate line drawings from 3D models. Our architecture incorporates a differentiable module operating on geometric features of the 3D model, and an image-based module operating on view-based shape representations. At test time, geometric and view-based reasoning are combined with the help of a neural module to create a line drawing. The model is trained on a large number of crowdsourced comparisons of line drawings. Experiments demonstrate that our method achieves significant improvements in line drawing over the state-of-the-art when evaluated on standard benchmarks, resulting in drawings that are comparable to those produced by experienced human artists.

Abstract We present a method that detects boundaries of parts in 3D shapes represented as point clouds. Our method is based on a graph convolutional network architecture that outputs a probability for a point to lie in an area that separates two or more parts in a 3D shape. Our boundary detector is quite generic: it can be trained to localize boundaries of semantic parts or geometric primitives commonly used in 3D modeling. Our experiments demonstrate that our method can extract more accurate boundaries that are closer to ground‐truth ones compared to alternatives. We also demonstrate an application of our network to fine‐grained semantic shape segmentation, where we also show improvements in terms of part labeling performance.

We present a method that generates expressive talking-head videos from a single facial image with audio as the only input. In contrast to previous attempts to learn direct mappings from audio to raw pixels for creating talking faces, our method first disentangles the content and speaker information in the input audio signal. The audio content robustly controls the motion of lips and nearby facial regions, while the speaker information determines the specifics of facial expressions and the rest of the talking-head dynamics. Another key component of our method is the prediction of facial landmarks reflecting the speaker-aware dynamics. Based on this intermediate representation, our method works with many portrait images in a single unified framework, including artistic paintings, sketches, 2D cartoon characters, Japanese mangas, and stylized caricatures. In addition, our method generalizes well for faces and characters that were not observed during training. We present extensive quantitative and qualitative evaluation of our method, in addition to user studies, demonstrating generated talking-heads of significantly higher quality compared to prior state-of-the-art methods.

We present RigNet , an end-to-end automated method for producing animation rigs from input character models. Given an input 3D model representing an articulated character, RigNet predicts a skeleton that matches the animator expectations in joint placement and topology. It also estimates surface skin weights based on the predicted skeleton. Our method is based on a deep architecture that directly operates on the mesh representation without making assumptions on shape class and structure. The architecture is trained on a large and diverse collection of rigged models, including their mesh, skeletons and corresponding skin weights. Our evaluation is three-fold: we show better results than prior art when quantitatively compared to animator rigs; qualitatively we show that our rigs can be expressively posed and animated at multiple levels of detail; and finally, we evaluate the impact of various algorithm choices on our output rigs. 1

User generated 3D shapes in online repositories contain rich information about surfaces, primitives, and their geometric relations, often arranged in a hierarchy. We present a framework for learning representations of 3D shapes that reflect the information present in this meta data and show that it leads to improved generalization for semantic segmentation tasks. Our approach is a point embedding network that generates a vectorial representation of the 3D points such that it reflects the grouping hierarchy and tag data. The main challenge is that the data is noisy and highly variable. To this end, we present a tree-aware metric-learning approach and demonstrate that such learned embeddings offer excellent transfer to semantic segmentation tasks, especially when training data is limited. Our approach reduces the relative error by 10.2% with 8 training examples, by 11.72% with 120 training examples on the ShapeNet semantic segmentation benchmark, in comparison to the network trained from scratch. By utilizing tag data the relative error is reduced by 12.8% with 8 training examples, in comparison to the network trained from scratch. These improvements come at no additional labeling cost as the meta data is freely available.

We present a learning method for predicting animation skeletons for input 3D models of articulated characters. In contrast to previous approaches that fit pre-defined skeleton templates or predict fixed sets of joints, our method produces an animation skeleton tailored for the structure and geometry of the input 3D model. Our architecture is based on a stack of hourglass modules trained on a large dataset of 3D rigged characters mined from the web. It operates on the volumetric representation of the input 3D shapes augmented with geometric shape features that provide additional cues for joint and bone locations. Our method also enables intuitive user control of the level-of-detail for the output skeleton. Our evaluation demonstrates that our approach predicts animation skeletons that are much more similar to the ones created by humans compared to several alternatives and baselines.

In this paper we propose a neural message passing approach to augment an input 3D indoor scene with new objects matching their surroundings. Given an input, potentially incomplete, 3D scene and a query location, our method predicts a probability distribution over object types that fit well in that location. Our distribution is predicted though passing learned messages in a dense graph whose nodes represent objects in the input scene and edges represent spatial and structural relationships. By weighting messages through an attention mechanism, our method learns to focus on the most relevant surrounding scene context to predict new scene objects. We found that our method significantly outperforms state-of-the-art approaches in terms of correctly predicting objects missing in a scene based on our experiments in the SUNCG dataset. We also demonstrate other applications of our method, including context-based 3D object recognition and iterative scene generation.

Material understanding is critical for design, geometric modeling, and analysis of functional objects. We enable material-aware 3D shape analysis by employing a projective convolutional neural network architecture to learn material-aware descriptors from view-based representations of 3D points for point-wise material classification or material-aware retrieval. Unfortunately, only a small fraction of shapes in 3D repositories are labeled with physical materials, posing a challenge for learning methods. To address this challenge, we crowdsource a dataset of 3080 3D shapes with part-wise material labels. We focus on furniture models which exhibit interesting structure and material variability. In addition, we also contribute a high-quality expert-labeled benchmark of 115 shapes from Herman-Miller and IKEA for evaluation. We further apply a mesh-aware conditional random field, which incorporates rotational and reflective symmetries, to smooth our local material predictions across neighboring surface patches. We demonstrate the effectiveness of our learned descriptors for automatic texturing, material-aware retrieval, and physical simulation.

We present a neural architecture that takes as input a 2D or 3D shape and outputs a program that generates the shape. The instructions in our program are based on constructive solid geometry principles, i.e., a set of boolean operations on shape primitives defined recursively. Bottom-up techniques for this shape parsing task rely on primitive detection and are inherently slow since the search space over possible primitive combinations is large. In contrast, our model uses a recurrent neural network that parses the input shape in a top-down manner, which is significantly faster and yields a compact and easy-to-interpret sequence of modeling instructions. Our model is also more effective as a shape detector compared to existing state-of-the-art detection techniques. We finally demonstrate that our network can be trained on novel datasets without ground-truth program annotations through policy gradient techniques.

We present a network architecture for processing point clouds that directly operates on a collection of points represented as a sparse set of samples in a high-dimensional lattice. Naively applying convolutions on this lattice scales poorly, both in terms of memory and computational cost, as the size of the lattice increases. Instead, our network uses sparse bilateral convolutional layers as building blocks. These layers maintain efficiency by using indexing structures to apply convolutions only on occupied parts of the lattice, and allow flexible specifications of the lattice structure enabling hierarchical and spatially-aware feature learning, as well as joint 2D-3D reasoning. Both point-based and image-based representations can be easily incorporated in a network with such layers and the resulting model can be trained in an end-to-end manner. We present results on 3D segmentation tasks where our approach outperforms existing state-of-the-art techniques.

We present a new local descriptor for 3D shapes, directly applicable to a wide range of shape analysis problems such as point correspondences, semantic segmentation, affordance prediction, and shape-to-scan matching. The descriptor is produced by a convolutional network that is trained to embed geometrically and semantically similar points close to one another in descriptor space. The network processes surface neighborhoods around points on a shape that are captured at multiple scales by a succession of progressively zoomed-out views, taken from carefully selected camera positions. We leverage two extremely large sources of data to train our network. First, since our network processes rendered views in the form of 2D images, we repurpose architectures pretrained on massive image datasets. Second, we automatically generate a synthetic dense point correspondence dataset by nonrigid alignment of corresponding shape parts in a large collection of segmented 3D models. As a result of these design choices, our network effectively encodes multiscale local context and fine-grained surface detail. Our network can be trained to produce either category-specific descriptors or more generic descriptors by learning from multiple shape categories. Once trained, at test time, the network extracts local descriptors for shapes without requiring any part segmentation as input. Our method can produce effective local descriptors even for shapes whose category is unknown or different from the ones used while training. We demonstrate through several experiments that our learned local descriptors are more discriminative compared to state-of-the-art alternatives and are effective in a variety of shape analysis applications.

Object functionality is often expressed through part articulation - as when the two rigid parts of a scissor pivot against each other to perform the cutting function. Such articulations are often similar across objects within the same functional category. In this paper we explore how the observation of different articulation states provides evidence for part structure and motion of 3D objects. Our method takes as input a pair of unsegmented shapes representing two different articulation states of two functionally related objects, and induces their common parts along with their underlying rigid motion. This is a challenging setting, as we assume no prior shape structure, no prior shape category information, no consistent shape orientation, the articulation states may belong to objects of different geometry, plus we allow inputs to be noisy and partial scans, or point clouds lifted from RGB images. Our method learns a neural network architecture with three modules that respectively propose correspondences, estimate 3D deformation flows, and perform segmentation. To achieve optimal performance, our architecture alternates between correspondence, deformation flow, and segmentation prediction iteratively in an ICP-like fashion. Our results demonstrate that our method significantly outperforms state-of-the-art techniques in the task of discovering articulated parts of objects. In addition, our part induction is object-class agnostic and successfully generalizes to new and unseen objects.

We present a novel deep-learning based approach to producing animator-centric speech motion curves that drive a JALI or standard FACS-based production face-rig, directly from input audio. Our three-stage Long Short-Term Memory (LSTM) network architecture is motivated by psycho-linguistic insights: segmenting speech audio into a stream of phonetic-groups is sufficient for viseme construction; speech styles like mumbling or shouting are strongly co-related to the motion of facial landmarks; and animator style is encoded in viseme motion curve profiles. Our contribution is an automatic real-time lip-synchronization from audio solution that integrates seamlessly into existing animation pipelines. We evaluate our results by: cross-validation to ground-truth data; animator critique and edits; visual comparison to recent deep-learning lip-synchronization solutions; and showing our approach to be resilient to diversity in speaker and language.

We propose a method for reconstructing 3D shapes from 2D sketches in the form of line drawings. Our method takes as input a single sketch, or multiple sketches, and outputs a dense point cloud representing a 3D reconstruction of the input sketch(es). The point cloud is then converted into a polygon mesh. At the heart of our method lies a deep, encoder-decoder network. The encoder converts the sketch into a compact representation encoding shape information. The decoder converts this representation into depth and normal maps capturing the underlying surface from several output viewpoints. The multi-view maps are then consolidated into a 3D point cloud by solving an optimization problem that fuses depth and normals across all viewpoints. Based on our experiments, compared to other methods, such as volumetric networks, our architecture offers several advantages, including more faithful reconstruction, higher output surface resolution, better preservation of topology and shape structure.

This paper introduces a deep architecture for segmenting 3D objects into their labeled semantic parts. Our architecture combines image-based Fully Convolutional Networks (FCNs) and surface-based Conditional Random Fields (CRFs) to yield coherent segmentations of 3D shapes. The image-based FCNs are used for efficient view-based reasoning about 3D object parts. Through a special projection layer, FCN outputs are effectively aggregated across multiple views and scales, then are projected onto the 3D object surfaces. Finally, a surface-based CRF combines the projected outputs with geometric consistency cues to yield coherent segmentations. The whole architecture (multi-view FCNs and CRF) is trained end-to-end. Our approach significantly outperforms the existing state-of-the-art methods in the currently largest segmentation benchmark (ShapeNet). Finally, we demonstrate promising segmentation results on noisy 3D shapes acquired from consumer-grade depth cameras.

We propose a data-driven method for recovering missing parts of 3D shapes. Our method is based on a new deep learning architecture consisting of two sub-networks: a global structure inference network and a local geometry refinement network. The global structure inference network incorporates a long short-term memorized context fusion module (LSTM-CF) that infers the global structure of the shape based on multi-view depth information provided as part of the input. It also includes a 3D fully convolutional (3DFCN) module that further enriches the global structure representation according to volumetric information in the input. Under the guidance of the global structure network, the local geometry refinement network takes as input local 3D patches around missing regions, and progressively produces a high-resolution, complete surface through a volumetric encoder-decoder architecture. Our method jointly trains the global structure inference and local geometry refinement networks in an end-to-end manner. We perform qualitative and quantitative evaluations on six object categories, demonstrating that our method outperforms existing state-of-the-art work on shape completion.

Abstract Data‐driven methods serve an increasingly important role in discovering geometric, structural and semantic relationships between shapes. In contrast to traditional approaches that process shapes in isolation of each other, data‐driven methods aggregate information from 3D model collections to improve the analysis, modelling and editing of shapes. Data‐driven methods are also able to learn computational models that reason about properties and relationships of shapes without relying on hard‐coded rules or explicitly programmed instructions. Through reviewing the literature, we provide an overview of the main concepts and components of these methods, as well as discuss their application to classification, segmentation, matching, reconstruction, modelling and exploration, as well as scene analysis and synthesis. We conclude our report with ideas that can inspire future research in data‐driven shape analysis and processing.

We introduce a deep learning approach for grouping discrete patterns common in graphical designs. Our approach is based on a convolutional neural network architecture that learns a grouping measure defined over a pair of pattern elements. Motivated by perceptual grouping principles, the key feature of our network is the encoding of element shape, context, symmetries, and structural arrangements. These element properties are all jointly considered and appropriately weighted in our grouping measure. To better align our measure with human perceptions for grouping, we train our network on a large, human-annotated dataset of pattern groupings consisting of patterns at varying granularity levels, with rich element relations and varieties, and tempered with noise and other data imperfections. Experimental results demonstrate that our deep-learned measure leads to robust grouping results.

Procedural modeling techniques can produce high quality visual content through complex rule sets. However, controlling the outputs of these techniques for design purposes is often notoriously difficult for users due to the large number of parameters involved in these rule sets and also their non-linear relationship to the resulting content. To circumvent this problem, we present a sketch-based approach to procedural modeling. Given an approximate and abstract hand-drawn 2D sketch provided by a user, our algorithm automatically computes a set of procedural model parameters, which in turn yield multiple, detailed output shapes that resemble the user's input sketch. The user can then select an output shape, or further modify the sketch to explore alternative ones. At the heart of our approach is a deep Convolutional Neural Network (CNN) that is trained to map sketches to procedural model parameters. The network is trained by large amounts of automatically generated synthetic line drawings. By using an intuitive medium, i.e., freehand sketching as input, users are set free from manually adjusting procedural model parameters, yet they are still able to create high quality content. We demonstrate the accuracy and efficacy of our method in a variety of procedural modeling scenarios including design of man-made and organic shapes.

Abstract Recently there has been an increasing demand for software that can help designers create functional 3D objects with required physical strength. We introduce a generic and extensible method that directly optimizes a shape subject to physical and geometric constraints. Given an input shape, our method optimizes directly its input mesh representation until it can withstand specified external forces, while remaining similar to the original shape. Our method performs physics simulation and shape optimization together in a unified framework, where the physics simulator is an integral part of the optimizer. We employ geometric constraints to preserve surface details and shape symmetry, and adapt a second‐order method with analytic gradients to improve convergence and computation time. Our method provides several advantages over previous work, including the ability to handle general shape deformations, preservation of surface details, and incorporation of user‐defined constraints. We demonstrate the effectiveness of our method on a variety of prinTable 3D objects through detailed simulations as well as physical validations.

When geometric models with a desired combination of style and functionality are not available, they currently need to be created manually. We facilitate algorithmic synthesis of 3D models of man-made shapes which combines user-specified style, described via anexemplarshape, and functionality, encoded by a functionally differenttargetshape. Our method automatically transfers the style of the exemplar to the target, creating the desired combination. The main challenge in performing cross-functional style transfer is to implicitly separate an object's style from its function: while stylistically the output shapes should be as close as possible to the exemplar, their original functionality and structure, as encoded by the target, should be strictly preserved. Recent literature point to the presence of similarly shaped, salientgeometric elementsas a main indicator of stylistic similarity between 3D shapes. We therefore transfer the exemplar style to the target via a sequence of element-level operations. We allow onlycompatibleoperations, ones that do not affect the target functionality. To this end, we introduce a cross-structural element compatibility metric that estimates the impact of each operation on the edited shape. Our metric is based on the global context and coarse geometry of evaluated elements, and is trained on databases of 3D objects. We use this metric to cast style transfer as a tabu search, which incrementally updates the target shape using compatible operations, progressively increasing its style similarity to the exemplar while strictly maintaining its functionality at each step. We evaluate our framework across a range of man-made objects including furniture, light fixtures, and tableware, and perform a number of user studies confirming that it produces convincing outputs combining the desired style and function.

A longstanding question in computer vision concerns the representation of 3D shapes for recognition: should 3D shapes be represented with descriptors operating on their native 3D formats, such as voxel grid or polygon mesh, or can they be effectively represented with view-based descriptors? We address this question in the context of learning to recognize 3D shapes from a collection of their rendered views on 2D images. We first present a standard CNN architecture trained to recognize the shapes' rendered views independently of each other, and show that a 3D shape can be recognized even from a single view at an accuracy far higher than using state-of-the-art 3D shape descriptors. Recognition rates further increase when multiple views of the shapes are provided. In addition, we present a novel CNN architecture that combines information from multiple views of a 3D shape into a single and compact shape descriptor offering even better recognition performance. The same architecture can be applied to accurately recognize human hand-drawn sketches of shapes. We conclude that a collection of 2D views can be highly informative for 3D shape recognition and is amenable to emerging CNN architectures and their derivatives.

Abstract We present a method for joint analysis and synthesis of geometrically diverse 3D shape families. Our method first learns part‐based templates such that an optimal set of fuzzy point and part correspondences is computed between the shapes of an input collection based on a probabilistic deformation model. In contrast to previous template‐based approaches, the geometry and deformation parameters of our part‐based templates are learned from scratch. Based on the estimated shape correspondence, our method also learns a probabilistic generative model that hierarchically captures statistical relationships of corresponding surface point positions and parts as well as their existence in the input shapes. A deep learning procedure is used to capture these hierarchical relationships. The resulting generative model is used to produce control point arrangements that drive shape synthesis by combining and deforming parts from the input collection. The generative model also yields compact shape descriptors that are used to perform fine‐grained classification. Finally, it can be also coupled with the probabilistic deformation model to further improve shape correspondence. We provide qualitative and quantitative evaluations of our method for shape correspondence, segmentation, fine‐grained classification and synthesis. Our experiments demonstrate superior correspondence and segmentation results than previous state‐of‐the‐art approaches.

The human perception of stylistic similarity transcends structure and function: for instance, a bed and a dresser may share a common style. An algorithmically computed style similarity measure that mimics human perception can benefit a range of computer graphics applications. Previous work in style analysis focused on shapes within the same class, and leveraged structural similarity between these shapes to facilitate analysis. In contrast, we introduce the firststructure-transcendingstyle similarity measure and validate it to be well aligned with human perception of stylistic similarity. Our measure is inspired by observations about style similarity in art history literature, which point to the presence of similarly shaped, salient,geometric elementsas one of the key indicators of stylistic similarity. We translate these observations into an algorithmic measure by first quantifying the geometric properties that make humans perceive geometric elements as similarly shaped and salient in the context of style, then employing this quantification to detect pairs of matching style related elements on the analyzed models, and finally collating the element-level geometric similarity measurements into an object-level style measure consistent with human perception. To achieve this consistency we employ crowdsourcing to quantify the different components of our measure; we learn the relative perceptual importance of a range of elementary shape distances and other parameters used in our measurement from 50K responses to cross-structure style similarity queries provided by over 2500 participants.We train and validate our method on this dataset, showing it to successfully predict relative style similarity with near 90% accuracy based on 10-fold cross-validation.

, a mobile solution that leverages a wristband containing a 9-axis inertial measurement unit to capture changes in the orientation of a person's arm, and a machine learning pipeline that processes this data to accurately detect smoking gestures and sessions in real-time. Our key innovations are fourfold: a) an arm trajectory-based method that extracts candidate hand-to-mouth gestures, b) a set of trajectory-based features to distinguish smoking gestures from confounding gestures including eating and drinking, c) a probabilistic model that analyzes sequences of hand-to-mouth gestures and infers which gestures are part of individual smoking sessions, and d) a method that leverages multiple IMUs placed on a person's body together with 3D animation of a person's arm to reduce burden of self-reports for labeled data collection. Our experiments show that our gesture recognition algorithm can detect smoking gestures with high accuracy (95.7%), precision (91%) and recall (81%). We also report a user study that demonstrates that we can accurately detect the number of smoking sessions with very few false positives over the period of a day, and that we can reliably extract the beginning and end of smoking session periods.

Abstract Style transfer aims to apply the style of an exemplar model to a target one, while retaining the target's structure. The main challenge in this process is to algorithmically distinguish style from structure, a high‐level, potentially ill‐posed cognitive task. Inspired by cognitive science research we recast style transfer in terms ofshape analogies. In IQ testing, shape analogy queries present the subject with three shapes:source, targetandexemplar, and ask them to select anoutputsuch that the transformation, oranalogy, from the exemplar to the output is similar to that from the source to the target. The logical process involved in identifying the source‐to‐target analogies implicitly detects the structural differences between the source and target and can be used effectively to facilitate style transfer. Since the exemplar has a similar structure to the source, applying the analogy to the exemplar will provide the output we seek. The main technical challenge we address is to compute the source to target analogies, consistent with human logic. We observe that the typical analogies we look for consist of a small set of simple transformations, which when applied to the exemplar generate a continuous, seamless output model. To assemble a shape analogy, we compute an optimal set of source‐to‐target transformations, such that the assembled analogy best fits these criteria. The assembled analogy is then applied to the exemplar shape to produce the desired output model. We use the proposed framework to seamlessly transfer a variety of style properties between 2D and 3D objects and demonstrate significant improvements over the state of the art in style transfer. We further show that our framework can be used to successfully complete partial scans with the help of a user provided structural template, coherently propagating scan style across the completed surfaces.

We present AttribIt, an approach for people to create visual content using relative semantic attributes expressed in linguistic terms. During an off-line processing step, AttribIt learns semantic attributes for design components that reflect the high-level intent people may have for creating content in a domain (e.g. adjectives such as "dangerous", "scary" or "strong") and ranks them according to the strength of each learned attribute. Then, during an interactive design session, a person can explore different combinations of visual components using commands based on relative attributes (e.g. "make this part more dangerous"). Novel designs are assembled in real-time as the strengths of selected attributes are varied, enabling rapid, in-situ exploration of candidate designs. We applied this approach to 3D modeling and web design. Experiments suggest this interface is an effective alternative for novices performing tasks with high-level design goals.

Abstract We present a novel particle‐based method for stable simulation of elasto‐plastic materials. The main contribution of our method is an implicit numerical integrator, using a physically‐based model, for computing particles that undergo both elastic and plastic deformations. The main advantage of our implicit integrator is that it allows the use of large time steps while still preserving stable and physically plausible simulation results. As a key component of our algorithm, at each time step we compute the particle positions and velocities based on a sparse linear system, which we solve efficiently on the graphics hardware. Compared to existing techniques, our method allows for a much wider range of stiffness and plasticity settings. In addition, our method can significantly reduce the computation cost for certain range of material types. We demonstrate fast and stable simulations for a variety of elasto‐plastic materials, ranging from highly stiffelastic materials to highly plastic ones.

We present an approach to synthesizing shapes from complex domains, by identifying new plausible combinations of components from existing shapes. Our primary contribution is a new generative model of component-based shape structure. The model represents probabilistic relationships between properties of shape components, and relates them to learned underlying causes of structural variability within the domain. These causes are treated as latent variables, leading to a compact representation that can be effectively learned without supervision from a set of compatibly segmented shapes. We evaluate the model on a number of shape datasets with complex structural variability and demonstrate its application to amplification of shape databases and to interactive shape synthesis.

This article presents an algorithm for learning hatching styles from line drawings. An artist draws a single hatching illustration of a 3D object. Her strokes are analyzed to extract the following per-pixel properties: hatching level (hatching, cross-hatching, or no strokes), stroke orientation, spacing, intensity, length, and thickness. A mapping is learned from input geometric, contextual, and shading features of the 3D object to these hatching properties, using classification, regression, and clustering techniques. Then, a new illustration can be generated in the artist's style, as follows. First, given a new view of a 3D object, the learned mapping is applied to synthesize target stroke properties for each pixel. A new illustration is then generated by synthesizing hatching strokes according to the target properties.

Assembly-based modeling is a promising approach to broadening the accessibility of 3D modeling. In assembly-based modeling, new models are assembled from shape components extracted from a database. A key challenge in assembly-based modeling is the identification of relevant components to be presented to the user. In this paper, we introduce a probabilistic reasoning approach to this problem. Given a repository of shapes, our approach learns a probabilistic graphical model that encodes semantic and geometric relationships among shape components. The probabilistic model is used to present components that are semantically and stylistically compatible with the 3D model that is being assembled. Our experiments indicate that the probabilistic model increases the relevance of presented components.

This paper presents a data-driven approach to simultaneous segmentation and labeling of parts in 3D meshes. An objective function is formulated as a Conditional Random Field model, with terms assessing the consistency of faces with labels, and terms between labels of neighboring faces. The objective function is learned from a collection of labeled training meshes. The algorithm uses hundreds of geometric and contextual label features and learns different types of segmentations for different tasks, without requiring manual parameter tuning. Our algorithm achieves a significant improvement in results over the state-of-the-art when evaluated on the Princeton Segmentation Benchmark, often producing segmentations and labelings comparable to those produced by humans.

This paper presents a method for estimating geographic location for sequences of time-stamped photographs. A prior distribution over travel describes the likelihood of traveling from one location to another during a given time interval. This distribution is based on a training database of 6 million photographs from Flickr.com. An image likelihood for each location is defined by matching a test photograph against the training database. Inferring location for images in a test sequence is then performed using the Forward-Backward algorithm, and the model can be adapted to individual users as well. Using temporal constraints allows our method to geolocate images without recognizable landmarks, and images with no geographic cues whatsoever. This method achieves a substantial performance improvement over the best-available baseline, and geolocates some users' images with near-perfect accuracy.

Abstract We present a real‐time relighting and shadowing method for dynamic scenes with varying lighting, view and BRDFs. Our approach is based on a compact representation of reflectance data that allows for changing the BRDF at run‐time and a data‐driven method for accurately synthesizing self‐shadows on articulated and deformable geometries. Unlike previous self‐shadowing approaches, we do not rely on local blocking heuristics. We do not fit a model to the BRDF‐weighted visibility, but rather only to the visibility that changes during animation. In this manner, our model is more compact than previous techniques and requires less computation both during fitting and at run‐time. Our reflectance product operators can re‐integrate arbitrary low‐frequency view‐dependent BRDF effects on‐the‐fly and are compatible with all previous dynamic visibility generation techniques as well as our own data‐driven visibility model. We apply our reflectance product operators to three different visibility generation models, and our data‐driven model can achieve framerates well over 300Hz.

Abstract Shape segmentations designed for different applications show significant variation in the composition of their parts. In this paper, we introduce the segmentation and labeling of shape based on the simultaneous optimization of multiple heterogenous objectives that capture application‐specific segmentation criteria. We present a number of efficient objective functions that capture useful shape adjectives (compact, flat, narrow, perpendicular, etc.) Segmentation descriptions within our framework combine multiple such objective functions with optional labels to define each part. The optimization problem is simplified by proposing weighted Voronoi partitioning as a compact and continuous parametrization of spatially embedded shape segmentations. Separation of spatially close but geodesically distant parts is made possible using multi‐dimensional scaling prior to Voronoi partitioning. Optimization begins with an initial segmentation found using the centroids of a k‐means clustering of surface elements. This partition is automatically labeled to optimize heterogeneous part objectives and the Voronoi centers and their weights optimized using Generalized Pattern Search. We illustrate our framework using several diverse segmentation applications: consistent segmentations with semantic labels, bounding volume hierarchies for path tracing, and automatic rig and clothing transfer between animation characters.

This article presents a method for real-time line drawing of deforming objects. Object-space line drawing algorithms for many types of curves, including suggestive contours, highlights, ridges, and valleys, rely on surface curvature and curvature derivatives. Unfortunately, these curvatures and their derivatives cannot be computed in real-time for animated, deforming objects. In a preprocessing step, our method learns the mapping from a low-dimensional set of animation parameters (e.g., joint angles) to surface curvatures for a deforming 3D mesh. The learned model can then accurately and efficiently predict curvatures and their derivatives, enabling real-time object-space rendering of suggestive contours and other such curves. This represents an order-of-magnitude speedup over the fastest existing algorithm capable of estimating curvatures and their derivatives accurately enough for many different types of line drawings. The learned model can generalize to novel animation sequences and is also very compact, typically requiring a few megabytes of storage at runtime. We demonstrate our method for various types of animated objects, including skeleton-based characters, cloth simulation, and blend-shape facial animation, using a variety of nonphotorealistic rendering styles. An important component of our system is the use of dimensionality reduction for differential mesh data. We show that Independent Component Analysis (ICA) yields localized basis functions, and gives superior generalization performance to that of Principal Component Analysis (PCA).

A great challenge in information visualization today is to provide models and software that effectively integrate the graphics content of scenes with domain-specific knowledge so that the users can effectively query, interpret, personalize and manipulate the visualized information [1]. Moreover, it is important that the intelligent visualization applications are interoperable in the semantic web environment and thus, require that the models and software supporting them integrate state-of-the-art international standards for knowledge representation, graphics and multimedia. In this paper, we present a model, a methodology and a software framework for the semantic web (Intelligent 3D Visualization Platform - I3DVP) for the development of interoperable intelligent visualization applications that support the coupling of graphics and virtual reality scenes with domain knowledge of different domains. The graphics content and the semantics of the scenes are married into a consistent and cohesive ontological model while at the same time knowledge- based techniques for the querying, manipulation, and semantic personalization of the scenes are introduced. We also provide methods for knowledge driven information visualization and visualization- aided decision making based on inference by reasoning.